Article In Brief
Two children who were administered an antisense oligonucleotide (ASO) for a severe genetic epilepsy disorder developed hydrocephalus, and one of them died. Experts suggest that while the therapies have potential for these otherwise fatal disorders, they remain highly experimental, and greater efforts should be made to share data from other research and outcomes involving ASOs.
A 3-year-old girl treated with an antisense oligonucleotide (ASO) for a devastating form of genetically caused epileptic encephalopathy died last year after developing hydrocephalus, the principal investigator of the trial reported in October at the annual meeting of the American Neurological Association.
Another child treated with the same ASO also developed hydrocephalus but underwent a successful shunt placement, reported Timothy Yu, MD, PhD, a neurologist, associate professor of pediatrics at Harvard Medical School, and an attending physician in the division of genetics and genomics at Boston Children's Hospital.
Both children had a severe form of KCNT1 epileptic encephalopathy.
“Patients with KCNT1 epileptic encephalopathy can have a range of outcomes, but we specifically selected patients with early infantile migrating focal seizures, who have the absolute worst outcomes,” Dr. Yu said.
“In case series, half of these children die early, with a median age of death of 3 years of age; those that survive continue having dozens of seizures per day and suffer profound neurologic impairment. This severity was critical to our (and our institutional review board's) decision-making in this project.”
Dr. Yu previously revealed that the very first child he treated with an ASO, described in a paper published in 2019 in the New England Journal of Medicine, had also died—not because of an adverse event, but because of her disorder's effects.
The symptoms in all three girls had initially showed significant improvements. Despite the serious adverse events, Dr. Yu told Neurology Today that he remains committed to the study of ASOs for the treatment of neurologic disorders. Other neurologists and neuroscientists said that they, too, continue to have high hopes for ASOs.
“I think the promise is high,” Dr. Yu said. “We were shocked to see the seizures respond as well as they did, especially in patient No. 2 [the one who survived]. And yet we know that we also incurred this serious adverse event in both cases. The intervention has the capability of producing transformative change. We just need more studies. I don't think we will begin to know how successful this approach will be, or not, until we get the numbers up.”
A measure of how promising scientists continue to view ASOs as a means of targeting neurologic disorders can be seen in the recent establishment of several non profit organizations dedicated to their development.
The n-Lorem Foundation seeks to develop and provide ASOs free for so-called nano-rare disorders, those affecting fewer than 30 individuals worldwide. Other newly formed groups include the St. Jude's Center for Experimental Therapeutics, led by J. Paul Taylor, MD, PhD; the Dutch Center for RNA Therapeutics, led by Annemieke Aartsma-Rus, PhD; and the N=1 Collaborative. They seek to not only develop new ASOs but also foster data and share expertise among investigators worldwide.
Still, these serious adverse events have reinforced the view that use of ASOs for nano-rare disorders, when only a handful of children might receive them, places unique burdens on investigators.
“The fact that you can develop an ASO is only one part of the equation,” said Dr. Taylor. “The other part is the appropriateness of treating the patient. Do they have a phenotype that can reasonably expect to benefit from the intervention? Do the parents have realistic expectations? Will insurance cover it? How do we do this in a way that is not haphazard? How do we work cooperatively with people doing similar work? I do worry that this field needs to move forward carefully or we're going to repeat the errors of the gene therapy world.”
The Development of ASOs
ASOs, first developed over 20 years ago, are single strands of modified DNA-like molecules engineered to link to a complementary strand of messenger RNA. Rather than replacing a defective sequence, ASOs seek to squelch or restore the action of its target. And unlike in gene therapy, they are designed to link only to the target mRNA, without interfering with any other genetic activity.
On Sept. 19, 2016, the US Food and Drug Administration (FDA) announced approval of the first ASO for a neurologic disease, eterplirsen (Exondys 51). The injectable drug was developed by Sarepta Therapeutics for the treatment of Duchenne muscular dystrophy for patients with a confirmed mutation of the dystrophin gene amenable to exon 51 skipping, [Essentially, the ASO causes the RNA-editing function in the cell nucleus to skip the portion of the dystrophin gene, exon-51, that contains the mutation. It results in a shorter mRNA, which can be translated into a functional protein.]
Three months later, the FDA approved a second ASO, nusinersen (Spinraza), developed by Ionis Pharmaceuticals to treat spinal muscular atrophy, estimated to affect one in 10,000 live births.
“With the treatment, if started early, these kids can lead a normal life,” said Kenneth H. Fischbeck, MD, an NIH distinguished investigator at the National Institute of Neurological Disorders and Stroke. “I still get choked up every time I talk about it. It's such an amazing thing.”
“We were shocked to see the seizures respond as well as they did, especially in patient No. 2 [the one who survived]. And yet we know that we also incurred this serious adverse event in both cases. The intervention has the capability of producing transformative change. We just need more studies. I don't think we will begin to know how successful this approach will be, or not, until we get the numbers up.”—DR. TIMOTHY YU
The following year, Dr. Yu began work on another ASO, which he named milasen after the child for whom he developed it. Six-year-old Mila Makovec was battling a rare form of Batten disease caused by a mutation in the gene MFSD8, which causes neuronal ceroid lipofuscinosis type 7, an invariably progressive and fatal disease. In a paper published in the New England Journal of Medicine in October 2019, Dr. Yu and colleagues described how they had developed milasen within less than a year, finding no serious adverse events and a significant reduction in seizures as determined by serial EEGs and parental seizure diary.
Although the drug worked as intended, Mila had already developed progressive brain shrinkage prior to beginning the treatment, Dr. Yu told Neurology Today.
“She did really well the first two years,” he said. “Her seizures came down and stayed low and remained low in the third year. But the progressive brain shrinkage didn't stop. We and her family saw that she was becoming less responsive and losing her communication skills. At the beginning of her third year, the family decided she had been through enough. She stayed on the drug but moved to palliative care.”
Mila died on Feb. 11, 2021.
New Case Details
At the American Neurological Association (ANA) annual meeting in October, Dr. Yu presented an update on the ASO treatment he had offered to two young girls for KCNT1-associated epilepsy of infancy with migrating focal seizures. Children with this condition have very frequent daily seizures starting in early infancy that remain refractory to anti-seizure medications and have minimal to no milestone development. Often they die as young as age 3.
“Our group has been pursuing a pilot series of individualized ASO trials, each of them N-of-one or N-of-two,” he said.
The first girl to be treated, then age 2 ½, was having dozens of seizures per day that were uncontrolled with medication. Even so, she had no underlying metabolic disorder or structural brain abnormality, Dr. Yu reported.
After developing and testing numerous candidate ASOs, they settled on one that controlled the seizures in a mouse model of the disorder. They were granted an investigational new drug status by the FDA in September 2020 and began intrathecal injections every two weeks, beginning at 10 mg and ramping up to 20 mg at the second dose, 30 mg at the third dose, and 40 mg at the fourth.
“With the treatment, if started early, these kids can lead a normal life. I still get choked up every time I talk about it. It's such an amazing thing.”—DR. KENNETH H. FISHBECK
“We settled in at a maintenance dose initially of 40 mg, then went to 60 mg,” Dr. Yu said. “There were reductions in seizure frequency and severity over the first 100 days.”
The second patient, also age 2 ½, was enrolled a few months later. “The results were quite immediately apparent,” Dr. Yu said during his presentation at the ANA meeting. “At baseline her family had recorded five to 15 seizures daily. After the 20 mg dose, there was a slight downward trend. The trend continued at 30 mg. At 40, they really fell: zero to four per day. They stayed in this range for the next 50 days. It was a pretty striking response.”
Then, he said: “Now for the sad news. Both participants developed serious and arguably life-threatening events. In the eleventh month of the trial, on a routine study MRI, patient number one was found to have developed advanced hydrocephalus. Frankly it came as a shock. The patient had been doing well in the months leading up to the scan. We don't know exactly when it developed.”
The trial was paused, and after investigation and consideration of options, he reported, “the family withdrew from the trial, and there was a transition to comfort care.”
The second patient was also hospitalized after her fourth dose, initially for evaluation of dystonia. “She was found to have enlarging brain ventricles on serial scans, indicative of early stage hydrocephalus.” Dr. Yu said. “She underwent placement of a shunt, which resolved the hydrocephalus and led to successful discharge from the hospital.”
He noted that hydrocephalus had not been seen in any of the preclinical animal studies, nor in preclinical animal studies of other FDA-approved or investigational ASOs. “We inquired with other researchers,” Dr. Yu said at the meeting. “No one else had seen this.”
His group did notice, however, that a cohort of patients in a clinical trial of an ASO developed for Huntington's disease by Ionis were reported in 2020 to have developed dose-dependent ventricular enlargement.
“Our hypothesis is that we may have uncovered a new risk of ASO,” Dr. Yu said at the meeting. Perhaps, he said, the reaction occurred due to an overly aggressive treatment schedule.
“Case No. 2 has been monitored for a year,” he said. “We're considering a dose reduction and resuming treatment.”
The Question of Data Sharing
In late October, just days after Dr. Yu had given his report at the ANA meeting, approximately 50 experts in ASOs and the development of treatments for rare diseases attended a meeting on the subject at St. Jude. Dr. Fischbeck moderated a session on the importance of data sharing, particularly in trials involving only one or a handful of patients.
“There's an awful lot of scientific research that never gets published but that could be very helpful. There is great potential with these ASOs, but we need to remember that these are experimental treatments. We're learning that the human response is different from what was expected based on animal studies. It's very important that we're very cautious and that there is good reporting of results, both good and bad, as widely as possible.”—DR. JONATHAN MINK
“It came up during the session,” Dr. Fischbeck told Neurology Today, “whether the severe adverse events in Dr. Yu's trial could have been foreseen if more data about safety and efficacy had been shared by other investigators.”
Because Ionis had developed the ASO for Huntington's disease in which 88 cases of enlarged ventricles and three cases of frank hydrocephalus had been observed, he said, “Perhaps if they had been more forthcoming with their results in the Huntington's trial and the large amount of animal testing that they had done over the years, there is a chance it might have prevented these unfortunate adverse events.”
Ionis and the n-Lorem Foundation, Dr. Fischbeck said, “have a lot of data about not just this gene target but all the other gene targets they've studied,” Dr. Fischbeck said.
The founder and former leader of Ionis, who now leads the n-Lorem Foundation, told Neurology Today that both organizations have been “100 percent forthcoming.”
“We have offered to collaborate with everyone, including Tim,” said Stan Crooke, MD, PhD. “He, like every other investigator, has had multiple offers from us to collaborate. He chose for his own reasons to discover and develop this ASO himself and administer it without seeking our advice or participation.”
Dr. Yu objected to that characterization. “We've been talking to Stan's team for quite some time, ever since the original milasen,” Dr. Yu told Neurology Today. “For each of our programs, we've checked with a wide range of ASO and safety experts, including members of his Ionis team.”
During the session on data sharing, Dr. Fischbeck said, Dr. Crooke said he was willing to share the data and expertise that n-Lorem has access to.
“The question now is: will they be true to their word?” Dr. Fischbeck said. “What are the details going to be? But I think they're're opening the door to making the data available to investigators who want to put it to use.”
Jonathan Mink, MD, PhD, FAAN, the Frederick A. Horner M.D. Distinguished Professor in Pediatric Neurology and Chief of Child Neurology at University of Rochester Medical Center, said he agreed that data sharing is essential for the development of ASOs.
“We clinician-scientists are very quick to publish our successes and very slow to publish our failures,” Dr. Mink said. “There's an awful lot of scientific research that never gets published but that could be very helpful. There is great potential with these ASOs, but we need to remember that these are experimental treatments. We're learning that the human response is different from what was expected based on animal studies. It's very important that we're very cautious and that there is good reporting of results, both good and bad, as widely as possible.”
Dr. Yu noted that the N=1 collaborative is designed to foster exactly that kind of data sharing.